/
© 2026 RiffOn. All rights reserved.

Get your free personalized podcast brief

We scan new podcasts and send you the top 5 insights daily.

  1. Super Data Science: ML & AI Podcast with Jon Krohn
  2. 1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone
1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn · Sep 22, 2026

Katie Malone on reviving her podcast with AI, managing agents like people, and the paradoxes of data from Simpson's to Bainbridge's ironies.

AI Production Tools Revived Data Scientist Katie Malone's 'Linear Digressions' Podcast After 5.5 Years

AI significantly lowered the production burden of research and post-production, which was a major cause of burnout. This technological shift enabled Katie Malone, a solo creator, to relaunch her popular podcast that had been dormant for over five years.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago

The Core Skills for Managing People Directly Transfer to Managing Agentic AI

Defining tasks, setting acceptance criteria, handling ambiguity, and using a "trust but verify" approach are core management concepts. These skills apply equally to managing AI agents and human teams, reframing AI interaction from a purely technical skill to a managerial one.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago

Bainbridge's Irony: Better AI Erodes the Human Expertise Needed to Catch Its Failures

As automation becomes more reliable, humans trust it more and lose practice on routine tasks. This makes them less equipped to handle the complex, edge-case failures that the AI escalates, as their own skills have atrophied from lack of use.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago

Tom Davenport's "Process Slop" Describes a Future of AI Bots Exchanging Low-Quality Content

"Process Slop" occurs when automated systems on both sides of a transaction (e.g., a job applicant's AI and a recruiter's AI) communicate. This creates an automated but potentially meaningless exchange that removes human nuance and quality control from the process.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago

Benford's Law Uncovers Fraud by Spotting Unnaturally Uniform Leading Digits in Data

In natural datasets, '1' is the leading digit about 30% of the time. Fabricated data often shows an unnaturally even distribution of leading digits, which violates this statistical law and can be a strong signal of fraud for auditors and data scientists.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago

Shifting to AI Agent Management Disrupts the "Flow State" Prized by Engineers

The emerging role of an AI agent manager requires constant context switching, reviewing AI output, and articulating tasks. This is fundamentally different from the deep, focused work (flow state) of traditional coding, creating a psychological and cultural challenge for technical professionals.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago

Simpson's Paradox: UC Berkeley Data Falsely Showed Gender Bias Until Segmented by Department

Aggregate admissions data showed men were admitted at a higher rate. However, segmenting the data by department revealed the opposite: women tended to apply to more competitive programs with lower admission rates, which explained the misleading top-level trend.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago

Podcasting and Teaching Act as Forcing Functions for Deeply Learning Complex Topics

Committing to public outputs like podcast episodes or courses creates deadlines and an obligation to an audience. This pressure forces creators to learn subjects in detail—a strategy particularly useful for extroverts who may struggle with unstructured, solo study.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago

The Command-Line Terminal Is a Superior Environment for Achieving a "Flow State"

The focused, text-based nature of the terminal, free from the distractions of dragging and dropping in a GUI, helps technical users achieve a state of deep, uninterrupted work known as "flow." This psychological benefit is a key reason for its persistent preference among developers.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago

AI Detectors Can't Distinguish Between Thoughtful AI-Assisted Work and Low-Effort Slop

When AI writes the final text for a project based on deep human curation and intellectual framing, detection tools often flag it as 90% AI-generated. This conflates high-value, AI-assisted work with low-effort content, posing a challenge for creators and platforms.

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone thumbnail

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone

Super Data Science: ML & AI Podcast with Jon Krohn·12 days ago